Page 207
www.rsisinternational.org
ILEIID 2025 | International Journal of Research and Innovation in Social Science (IJRISS)
ISSN: 2454-6186 | DOI: 10.47772/IJRISS
Special Issue | Volume IX Issue XXII October 2025
Integrating ChatGPT into the Esquisse Task During Architectural
Learning Studios at the University
*
Megat Faridrullah bin Zolkefli
Architecture Program, Faculty of Built Environment, Universiti Teknologi MARA Cawangan Sarawak,
Kampus Samarahan
*Corresponding author
DOI: https://dx.doi.org/10.47772/IJRISS.2025.922ILEIID0020
Received: 22 September 2025; Accepted: 30 September 2025; Published: 22 October 2025
ABSTRACT
The esquisse, a quick design ideation exercise that tests students' ability to conceptualise, sketch, and
communicate design concepts in a short amount of time, is a key component of architectural education.
Although it is good at encouraging creativity and agility, many students struggle to come up with ideas,
articulate them, and get beyond creative obstacles, especially in the early years of study. This study
investigates how Universiti Teknologi MARA (UiTM) Sarawak's esquisse phase of studio learning
incorporates ChatGPT, a conversational artificial intelligence application. The goals are to evaluate how
ChatGPT contributes to better conceptual growth, a wider range of design concepts, and increased student self-
assurance and communication abilities. Instead of using AI as a tool for post-design analysis, the innovation is
in presenting it as an active ideation partner throughout the live design process. Students were able to improve
their narrative clarity, broaden their conceptual vocabulary, and get ideas for different solutions by having
reflective conversations with ChatGPT. Results show that design outcomes are more diverse, that ideation
anxiety is lower, and that critique engagement is higher. By encouraging more inclusive and flexible learning
environments and giving aspiring architects the tools they need to be creative and digitally literate, the study
benefits society. The strategy might develop into an AI-enhanced studio platform or Learning Management
System (LMS) plugin specifically for the architecture and design education markets from a commercialisation
standpoint. This places the work at the nexus of innovative education, AI technology, and creative pedagogy.
Keywords: ChatGPT, architectural education, esquisse, artificial intelligence, studio pedagogy
INTRODUCTION
It is commonly acknowledged that studio-based learning is a key component of architectural education, where
students hone their design abilities through iterative conceptualisation, sketching, and critiquing procedures.
The esquisse task, a quick design ideation activity that encourages students to come up with innovative
answers to design prompts in a limited amount of time, lies at the heart of this. Esquisse fosters spontaneity,
flexibility, and fast decision-makingskills that are critical to the practice of architecture.
The exercise itself frequently poses difficulties. Many students struggle to come up with ideas fast, articulate
them well, and remain confident when presenting their design concepts, especially in the early stages of their
studies. This "creative block" issue increases performance anxiety, decreases design diversity, and restricts
their capacity for critical discourse.
New avenues for enhancing creativity in design pedagogy have been made possible by recent advancements in
artificial intelligence (AI), particularly in conversational platforms like ChatGPT. Although artificial
intelligence (AI) has historically been used in post-design analysis, such as assessing structural efficiency or
environmental performance, its potential as an active ideation partner is still not fully understood. With its
Page 208
www.rsisinternational.org
ILEIID 2025 | International Journal of Research and Innovation in Social Science (IJRISS)
ISSN: 2454-6186 | DOI: 10.47772/IJRISS
Special Issue | Volume IX Issue XXII October 2025
conversational capabilities, ChatGPT provides a way to help students with the ideation process itself,
facilitating thoughtful discussion, vocabulary development, and conceptual growth.
In the architectural learning studios at Universiti Teknologi MARA (UiTM) Sarawak, this concept paper
suggests incorporating ChatGPT into the esquisse work. The study makes the case that this kind of integration
could improve design results, assist in overcoming student obstacles, and establish a new paradigm for AI-
enhanced architectural education.
LITERATURE REVIEW
The Esquisse in Architectural Education
The esquisse has long been a feature of architectural teaching, notably in European and Asian cultures. Schön
(1983) asserts that the quick sketch problem enables students to practise "reflection-in-action," which fosters
the capacity for simultaneous thought and design. Although esquisse exercises promote creativity and agility
(Anthony, 1991), a number of research show that students frequently struggle with conceptual fluency and
time constraint (Goldschmidt, 2014). Esquisse not designed as an educational platform for assessment. It is a
design tools primarily for creating 3D models.
Creativity Challenges in Studio Learning
According to other studies conducted in architectural studios, young students encounter "ideation block,"
which limits their ability to explore and forces them to rely on tried-and-true, safe solutions (Lawson, 2004).
Even though critique sessions are beneficial, students who have hazy concepts may become intimidated
(Anthony, 1991). This is in line with the requirement for encouraging scaffolding in design education, where
technology might be helpful.
AI in Design and Education
From generative design in engineering (Akin, 2010) to adaptive tutoring systems in education (Luckin et al.,
2016), the application of AI in creative fields has been expanding. However, post-design performance
analysis, including energy modelling and structural optimisation, is the main area of use for AI in architecture
education (Oxman, 2017). In the field of conceptual ideation support, very little has been investigated.
ChatGPT and Conversational AI in Learning
OpenAI's ChatGPT is a novel approach to AI communication using natural language processing. Research
indicates that it can improve reflective learning, increase vocabulary, and lower anxiety by providing
conversational assistance (Kasneci et al., 2023). ChatGPT's capacity to offer iterative feedback and encourage
alternative thinking offers a special opportunity in design education, where concept articulation is just as
crucial as generation.
DISCUSSION
ChatGPT's incorporation into the esquisse task sparks fresh discussions about how to strike a balance between
machine aid and human innovation in architecture education. The conversation must also cover the wider
ramifications for teaching, learning, and professional practice, even though the initial results point to beneficial
contributions to student ideation and communication.
Opportunities for Enriched Learning
The most obvious possibility is to encourage students' self-confidence in their creative abilities. The esquisse
frequently turns into a stressful task that reveals the design thinking weaknesses of inexperienced architecture
students. ChatGPT lessens this cognitive strain by offering instant verbal cues and different viewpoints.
Students are urged to think in different ways, which broadens their vocabulary of design terms.
Page 209
www.rsisinternational.org
ILEIID 2025 | International Journal of Research and Innovation in Social Science (IJRISS)
ISSN: 2454-6186 | DOI: 10.47772/IJRISS
Special Issue | Volume IX Issue XXII October 2025
Customising educational experiences is an additional potential. ChatGPT may interact with each student
continually, tailoring its responses to the degree of detail and inquiry style provided, in contrast to traditional
critiques where feedback is constrained by time and instructor availability. Quieter or less confident students
are given equal opportunities to engage in design conversations thanks to this customised scaffolding, which
improves inclusion.
The inclusion of ChatGPT also encourages digital literacy in artistic endeavours. As computational design,
digital modelling, and AI-driven simulations become more integrated into architectural practice, exposing
students to AI early on as an ideation partner helps them get ready for professional settings where technology
fosters creativity.
Challenges and Risks
Despite its promise, a number of risks need to be taken into account.
Over-reliance on AI: Students run the risk of becoming overly reliant on AI recommendations, which would
impair their capacity for independent thought. If ChatGPT is utilised carelessly, it may accidentally
homogenise ideas rather than inspire originality.
Output quality and applicability: AI-generated content can occasionally be imprecise, ambiguous, or culturally
unrelated. In the absence of critical analysis, students may include false ideas in their designs.
Authorship and ethical issues: It's unclear if concepts generated by AI qualify as original. It is necessary to
clarify ownership and academic integrity concerns if ChatGPT plays a substantial role in concept development.
Access equity: Not every student may have equal access to dependable online resources. Those with restricted
gadgets or internet may unintentionally be at a disadvantage if AI used excessively in teaching.
To overcome the over-reliance on AI, the approach for the esquisse task in the studio started without any AI
tools assistance. Students are required to produce their esquisse models and need to explain their understanding
on the product produced (Figure 1). After the presentations done, lecturers and another group of students will
be giving their feedback and input on what they see from the esquisse produced. The incorporation of
ChatGPT came after the task done where the student can see another point of view and helping them to
develop their ideas further.
Pedagogical Implications
By incorporating ChatGPT into the curriculum, the lecturer's position is radically altered. Lecturers can
refocus their attention on higher-order design critique, such as contextual integration, spatial quality, and
theoretical foundation, rather than merely facilitating initial ideas. Studio culture could be improved by this
shift, becoming less limited by early-stage creative barriers and more dialogical.
Additionally, ChatGPT might contribute to a more varied design discussion within criticisms. Students who
previously hesitated to submit partial ideas (Figure 2) but now arrive with more developed concepts (Figure 3)
make richer peer discussions and a more inclusive learning environment possible. However, this requires
cautious guidance: educators must encourage pupils to regard AI responses as “suggestive inputs” rather than
definitive solutions.
Future Directions
Beyond the esquisse, AI-enhanced design education holds great promise. ChatGPT may eventually be
included into studio platforms or LMS plugins that monitor student development, modify prompts in response
to performance, and even incorporate tools for visual or spatial design. Ideation may be further enhanced by
combining conversational AI with generative design software, which offers multimodal feedback (text + sketch
+ data).
Page 210
www.rsisinternational.org
ILEIID 2025 | International Journal of Research and Innovation in Social Science (IJRISS)
ISSN: 2454-6186 | DOI: 10.47772/IJRISS
Special Issue | Volume IX Issue XXII October 2025
In order to determine if frequent usage of ChatGPT enhances or detracts from independent creativity, future
research should incorporate longitudinal investigations. Studies that compare different institutions could also
determine whether curriculum and cultural factors have an impact on how well AI works in studio instruction.
Lastly, establishing ethical guidelines and best practices for integrating AI into creative fields would require
cooperation between educators, technologists, and legislators.
Figure 1: Students explaining their esquisse.
Figure 2: Initial Idea by the Student.
Page 211
www.rsisinternational.org
ILEIID 2025 | International Journal of Research and Innovation in Social Science (IJRISS)
ISSN: 2454-6186 | DOI: 10.47772/IJRISS
Special Issue | Volume IX Issue XXII October 2025
Figure 4: ChatGPT feedback as suggestive inputs
CONCLUSION
The use of ChatGPT into the esquisse job during architectural studio learning at UiTM Sarawak was suggested
in this concept paper. Although the esquisse is still a useful teaching tool for encouraging creativity, agility,
and communication, many students struggle to articulate their ideas, have limited conceptual range, and have
ideation anxiety.
This study proposes a new method of studio education that blends digital scaffolding with human creativity by
presenting ChatGPT as an active ideation partner. Research shows that although design studios encourage
introspection and experimentation, students frequently require extra assistance when creating concepts and
stories. At the same time, studies on AI in education show how technology may be used to reduce anxiety and
customise learning. This initiative offers a novel approach to design learning by occupying the nexus of these
domains.
More conceptual diversity, enhanced communication abilities, and less stress during ideation are among the
anticipated results. However, difficulties like an excessive dependence on AI, moral dilemmas, and digital
fairness issues need to be resolved with cautious direction and conscientious integration.
The study's importance ultimately rests in showing how AI may enhance conventional studio procedures rather
than take their place. By incorporating ChatGPT into the esquisse, studio culture may change to become a
more welcoming, adaptable, and technologically savvy setting. In addition to its immediate educational
advantages, the strategy has the potential to be developed into AI-enhanced studio platforms or LMS plugins,
which would broaden its influence into more commercial and instructional contexts.
Page 212
www.rsisinternational.org
ILEIID 2025 | International Journal of Research and Innovation in Social Science (IJRISS)
ISSN: 2454-6186 | DOI: 10.47772/IJRISS
Special Issue | Volume IX Issue XXII October 2025
To sum up, ChatGPT is a promising partner in architectural education that helps students get over obstacles to
creativity and emphasises the importance of the human imagination in design.
ACKNOWLEDGEMENTS
The author would like to express gratitude to Universiti Teknologi MARA (UiTM) Cawangan Sarawak for
providing the academic environment and support that made this study possible. Special thanks extended to the
architecture studio students in year 3 semester October 2024 who participated in the exploratory sessions and
offered valuable insights on the integration of ChatGPT into the esquisse task. Their openness to
experimentation and constructive feedback has been central to shaping this concept paper.
REFERENCES
1. Akin, Ö. (2010). Variants in design cognition. In J. S. Gero (Ed.), Design computing and cognition ’08 (pp.
105124). Springer. https://doi.org/10.1007/978-1-4020-8728-8_6
2. Anthony, K. H. (1991). Design juries on trial: The renaissance of the design studio. Van Nostrand
Reinhold.
3. Goldschmidt, G. (2014). Linkography: Unfolding the design process. MIT Press.
4. Kasneci, E., Sessler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., ... s & Kasneci, G.
(2023). ChatGPT for good? On opportunities and challenges of large language models for education.
Learning and Individual Differences, 103, 102274. https://doi.org/10.1016/j.lindif.2023.102274
5. Lawson, B. (2004). What designers know. Architectural Press.
6. Luckin, R., Holmes, W., Griffiths, M., & Forcier, L. B. (2016). Intelligence unleashed: An argument for AI
in education. Pearson.
7. Oxman, R. (2017). Thinking difference: Theories and models of parametric design thinking. Design Studies,
52, 439. https://doi.org/10.1016/j.destud.2017.06.001
8. Schön, D. A. (1983). The reflective practitioner: How professionals think in action. Basic Books.